Chips with everything makes for a hi-tech mess


If you think that your computer, being such a modern, hi-tech device, is — or surely must be — environmentally friendly, then think again.

Researchers at the United Nations University in Tokyo recently analyzed the material and energy inputs required to produce a 32-megabyte DRAM microchip, and what they discovered came as a shock. Their findings have attracted media attention worldwide.

Microchips are ubiquitous. They surround us in our daily lives, from the alarm clocks that wake us and the vehicles we ride in to work, to the cell phones, computers and other machines we depend on throughout the day.

Consciously or not, most of us probably assume that microchips are part of a swath of technological changes often labeled “dematerialization.” This is the notion that progress in technology offers radical cuts in the quantity of materials and energy needed to produce goods and services.

In 1998, Eric Williams and his colleagues at the U.N. University began investigating the production chain of silicon, the chief material used in making microchips. Two years later their research expanded to explore the larger environmental footprint of information technology as a whole.

Last month, the first phase of results from their work appeared in Environmental Science & Technology, a journal of the American Chemical Society.

At UNU, Williams works with UNU Special Adviser, Professor Robert Ayres and three part-time research assistants, Takuro Hatanaka, Huahong Gi and Kenichi Umeda. Their work is funded by the Japan Foundation Center for Global Partnership, the Takeda Foundation, the UNU/Institute of Advanced Studies and the Fulbright Foundation.

Williams and his colleagues didn’t set out to focus their research on the macro impacts of the microchip, but on a much broader front. “Historically, technological revolutions have had a huge effect on environmental issues, especially the combustion engine and electricity,” Williams said last week. “We realized that IT could also have substantial impacts and benefits for the environment, so we took it up as a research theme.

“The microchip research is part of the first phase, which is to examine the direct impacts associated with making, using and disposing of IT hardware. There are also many indirect impacts and benefits associated with how IT is used by business and society.”

So what are the environmental impacts of producing and using a 32-megabyte DRAM computer chip that weighs a mere 2 grams? The UNU team found that to make every one of the millions manufactured each year requires 32 kg of water, 1.6 kg of fossil fuels, 700 grams of elemental gases (mainly nitrogen), and 72 grams of chemicals (hundreds are used, including lethal arsine gas and corrosive hydrogen fluoride).

To make matters worse, Williams believes his findings are conservative. “We think the real numbers may be twice that,” he said, adding that rapid advances in technology aggravate the problem. “The fact that a chip has such a short lifespan, because the technology turns over so quickly, exacerbates the environmental impact.”

Additionally, the UNU article highlights the contrast in terms of “environmental weight” between microchips and traditional goods, noting on one hand that “secondary materials used in production [of microchips] total 630 times the mass of the final product. Compared with this, the researchers state that an automobile requires only about twice its weight in fossil fuels to produce — i.e. 1,500-3,000 kg.

Addressing the question of why the secondary use of materials is comparatively so high for semiconductor devices, the report’s authors note that, “Microchips and many other hi-tech goods are extremely low-entropy, highly organized forms of matter. Given that they are fabricated using relatively high-entropy starting materials, it is natural to expect that a substantial investment of energy and process materials is needed for the transformation into an organized form.”

Williams and his colleagues are delighted at the enthusiastic reaction to their research. “The response from the academic and general press has been tremendous. . . . Perhaps part of the reason for this is that the results are surprising and also accessible to a wide audience,” he noted.

However, the findings have given Williams pause for thought about the wider implications of microchips and the future of the IT field. “The main thing that strikes me is that making a microchip uses far more chemicals and energy than its small size belies, so much so that the micro-product has a macro-environmental impact. This also makes microchips and the computers that contain them qualitatively different from other appliances or products such as refrigerators or automobiles.”

In this latter respect, Williams again returned to the automobile comparison, saying, “The impacts of a car are primarily associated with the fuel used in operating the vehicle, not its production. For a computer the production phase is far more important, which changes the approach to managing environmental impacts.

“For a car, environmental performance is mostly about fuel-efficiency and emissions standards. For a computer, the main strategies are reduction of energy used in production and for consumers to maximize a machine’s lifetime. The former is a challenge for the private sector, while the latter requires awareness and cooperation from the general public.”

Clearly the UNU research has generated as many questions as it has answered. More answers — and infinitely more questions — are likely to emerge as Eric Williams and his colleagues move on to investigate the wider environmental effects of adopting IT.

“One concern is how IT-driven changes in lifestyles, and the spectrum of goods that consumers purchase, affects the energy-use associated with daily life. This needs to be understood to inform future policies addressing climate change. Also, a research program is under development with the aim of applying IT to the management of water and air-quality in industrializing countries,” explained Williams.

As for those of us who spend more time seeking convenience than researching it, the microchip offers a thoughtful lesson: Small may be beautiful, but beauty isn’t necessarily environmentally friendly.